Table 2 Coefficients from LPG primary versus secondary cooking fuel logistic regression model

From: Modelling of supply and demand-side determinants of liquefied petroleum gas consumption in peri-urban Cameroon, Ghana and Kenya

Coefficient

Estimate

Standard error

Test statistic (Z-score)

P value

Predicted probability (95% CI)

Intercept

1.30

0.55

2.36

0.03*

 

Country (referenced to Ghana)

    

0.69 (0.47, 0.85)

 Cameroon

–0.73

0.57

–1.26

0.21

0.55 (0.32, 0.77)

 Kenya

–0.79

0.66

–1.19

0.23

0.47 (0.24, 0.72)

Household income quartile (referenced to 1st quartile (lowest))

    

0.69 (0.47, 0.85)

 2nd quartile

0.12

0.17

0.70

0.48

0.71 (0.50, 0.86)

 3rd quartile

–0.01

0.20

–0.06

0.95

0.67 (0.46, 0.83)

 4th quartile (highest)

0.15

0.25

0.59

0.56

0.70 (0.49, 0.86)

Highest level of education in household

    

0.69 (0.47,0.85)

 Primary

0.20

0.34

0.59

0.55

0.79 (0.61, 0.90)

 Junior high

0.65

0.42

1.56

0.12

0.70 (0.51, 0.84)

 Secondary

0.80

0.32

2.48

0.01*

0.81 (0.66, 0.90)

 College or university

1.22

0.34

3.60

<0.001***

0.87 (0.74, 0.94)

Age (yr) of head of household (referenced to 18–24)

    

0.69 (0.47, 0.85)

 25–29

–0.09

0.16

–0.54

0.59

0.67 (0.46, 0.83)

 30–35

–0.22

0.16

–1.27

0.20

0.65 (0.43, 0.82)

 36–75

–0.37

0.15

–2.33

0.02*

0.61 (0.39, 0.79)

Sex of cooking fuel decision maker (referenced to female)

    

0.69 (0.47, 0.85)

 Male

0.55

0.15

3.61

<0.001***

0.79 (0.59, 0.90)

Number of household members (referenced to 1–2)

    

0.84 (0.68, 0.93)

 3–4

–0.87

0.21

–3.99

<0.001***

0.69 (0.47, 0.85)

 5–6

–1.57

0.22

–7.06

<0.001***

0.52 (0.31, 0.73)

 7+

–2.31

0.24

–9.52

<0.001***

0.35 (0.17, 0.58)

Number of children under 5 years old living in household (referenced to 1)

    

0.69 (0.47, 0.85)

 2

0.23

0.13

1.68

0.09

0.74 (0.53, 0.87)

 3–6

–0.19

0.16

–1.14

0.25

0.65 (0.42, 0.83)

Days cooking with LPG in previous week (referenced to 6 or less)

    

0.69 (0.47, 0.85)

 7 days (every day)

1.15

0.12

9.18

<0.001***

0.87 (0.73, 0.95)

Obtain any cooking fuels for free

    

0.69 (0.47,0.85)

 Yes

–1.31

0.17

–7.93

<0.001***

0.37 (0.18,0.60)

Transportation cost for obtaining LPG refill (referenced to <US$0.25)

    

0.69 (0.47, 0.85)

 US$0.26–0.50

–0.40

0.18

–2.24

0.02*

0.62 (0.40, 0.80)

 $0.51–1.00

–0.46

0.15

–2.96

0.003**

0.59 (0.36, 0.79)

 >US$1.00

–0.27

0.15

–1.72

0.08

0.63 (0.39, 0.81)

Travel time to obtain LPG refill (referenced to <10 min)

    

0.69 (0.47, 0.85)

 11–20 min

0.03

0.14

0.20

0.84

0.70 (0.48, 0.85)

 21–30 min

–0.35

0.15

–2.26

0.02*

0.61 (0.39, 0.80)

 >30 min

0.44

0.25

1.72

0.09

0.75 (0.54, 0.89)

LPG unavailable (referenced to often (more than once a month))

    

0.69 (0.47, 0.85)

 Sometimes (4–12 times a year)

0.03

0.28

0.11

0.91

0.68 (0.47, 0.84)

 Rarely (less than 4 times a year)

0.45

0.25

1.76

0.08

0.75 (0.56, 0.88)

 Never (always available)

0.84

0.24

3.75

<0.001***

0.84 (0.69, 0.92)

 Do not know/unsure

0.42

0.33

1.25

0.21

0.75 (0.54, 0.89)

Cost of refills (US$ kg–1) (referenced to <US$0.86 kg–1)

    

0.69 (0.47, 0.85)

 US$0.86–1.00 kg–1

–0.35

0.25

–1.40

0.20

0.60 (0.40, 0.80)

 US$1.01–1.10 kg–1

–0.74

0.28

–2.61

0.009**

0.52 (0.30, 0.73)

 >US$1.10 kg–1

–1.30

0.33

–3.88

<0.001***

0.40 (0.20, 0.64)

  1. n = 2,247.
  2. Continuous covariates mean-centred and categorical predictors held at the population proportion. Households that primarily used LPG indicated LPG being their main cooking fuel, whereas secondary LPG users were those that stated LPG was another cooking fuel they used aside from their main cooking fuel. P values were generated from two-sided t-tests. No adjustments were made for multiple comparisons. *P < 0.05, **P < 0.01, ***P < 0.001.
Back to article page